static int
+zpl_open(struct inode *ip, struct file *filp)
+{
+ cred_t *cr = CRED();
+ int error;
+
+ crhold(cr);
+ error = -zfs_open(ip, filp->f_mode, filp->f_flags, cr);
+ crfree(cr);
+ ASSERT3S(error, <=, 0);
+
+ if (error)
+ return (error);
+
+ return generic_file_open(ip, filp);
+}
+
+static int
+zpl_release(struct inode *ip, struct file *filp)
+{
+ cred_t *cr = CRED();
+ int error;
+
+ crhold(cr);
+ error = -zfs_close(ip, filp->f_flags, cr);
+ crfree(cr);
+ ASSERT3S(error, <=, 0);
+
+ return (error);
+}
+
+static int
zpl_readdir(struct file *filp, void *dirent, filldir_t filldir)
{
struct dentry *dentry = filp->f_path.dentry;
- cred_t *cr;
+ cred_t *cr = CRED();
int error;
- cr = (cred_t *)get_current_cred();
+ crhold(cr);
error = -zfs_readdir(dentry->d_inode, dirent, filldir,
&filp->f_pos, cr);
- put_cred(cr);
+ crfree(cr);
ASSERT3S(error, <=, 0);
return (error);
}
+/*
+ * 2.6.35 API change,
+ * As of 2.6.35 the dentry argument to the .fsync() vfs hook was deemed
+ * redundant. The dentry is still accessible via filp->f_path.dentry,
+ * and we are guaranteed that filp will never be NULL.
+ *
+ * 2.6.34 API change,
+ * Prior to 2.6.34 the nfsd kernel server would pass a NULL file struct *
+ * to the .fsync() hook. For this reason, we must be careful not to use
+ * filp unconditionally in the 3 argument case.
+ */
+#ifdef HAVE_2ARGS_FSYNC
+static int
+zpl_fsync(struct file *filp, int datasync)
+{
+ struct dentry *dentry = filp->f_path.dentry;
+#else
static int
zpl_fsync(struct file *filp, struct dentry *dentry, int datasync)
{
- cred_t *cr;
+#endif /* HAVE_2ARGS_FSYNC */
+ cred_t *cr = CRED();
int error;
- cr = (cred_t *)get_current_cred();
- error = -zfs_fsync(filp->f_path.dentry->d_inode, datasync, cr);
- put_cred(cr);
+ crhold(cr);
+ error = -zfs_fsync(dentry->d_inode, datasync, cr);
+ crfree(cr);
ASSERT3S(error, <=, 0);
return (error);
static ssize_t
zpl_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos)
{
- cred_t *cr;
+ cred_t *cr = CRED();
ssize_t read;
- cr = (cred_t *)get_current_cred();
+ crhold(cr);
read = zpl_read_common(filp->f_mapping->host, buf, len, *ppos,
UIO_USERSPACE, filp->f_flags, cr);
- put_cred(cr);
+ crfree(cr);
if (read < 0)
return (read);
static ssize_t
zpl_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos)
{
- cred_t *cr;
+ cred_t *cr = CRED();
ssize_t wrote;
- cr = (cred_t *)get_current_cred();
+ crhold(cr);
wrote = zpl_write_common(filp->f_mapping->host, buf, len, *ppos,
UIO_USERSPACE, filp->f_flags, cr);
- put_cred(cr);
+ crfree(cr);
if (wrote < 0)
return (wrote);
static int
zpl_mmap(struct file *filp, struct vm_area_struct *vma)
{
- znode_t *zp = ITOZ(filp->f_mapping->host);
+ struct inode *ip = filp->f_mapping->host;
+ znode_t *zp = ITOZ(ip);
int error;
+ error = -zfs_map(ip, vma->vm_pgoff, (caddr_t *)vma->vm_start,
+ (size_t)(vma->vm_end - vma->vm_start), vma->vm_flags);
+ if (error)
+ return (error);
+
error = generic_file_mmap(filp, vma);
if (error)
return (error);
return (error);
}
+static struct page **
+pages_vector_from_list(struct list_head *pages, unsigned nr_pages)
+{
+ struct page **pl;
+ struct page *t;
+ unsigned page_idx;
+
+ pl = kmalloc(sizeof(*pl) * nr_pages, GFP_NOFS);
+ if (!pl)
+ return ERR_PTR(-ENOMEM);
+
+ page_idx = 0;
+ list_for_each_entry_reverse(t, pages, lru) {
+ pl[page_idx] = t;
+ page_idx++;
+ }
+
+ return pl;
+}
+
+static int
+zpl_readpages(struct file *file, struct address_space *mapping,
+ struct list_head *pages, unsigned nr_pages)
+{
+ struct inode *ip;
+ struct page **pl;
+ struct page *p, *n;
+ int error;
+
+ ip = mapping->host;
+
+ pl = pages_vector_from_list(pages, nr_pages);
+ if (IS_ERR(pl))
+ return PTR_ERR(pl);
+
+ error = -zfs_getpage(ip, pl, nr_pages);
+ if (error)
+ goto error;
+
+ list_for_each_entry_safe_reverse(p, n, pages, lru) {
+
+ list_del(&p->lru);
+
+ flush_dcache_page(p);
+ SetPageUptodate(p);
+ unlock_page(p);
+ page_cache_release(p);
+ }
+
+error:
+ kfree(pl);
+ return error;
+}
+
/*
* Populate a page with data for the Linux page cache. This function is
* only used to support mmap(2). There will be an identical copy of the
zpl_readpage(struct file *filp, struct page *pp)
{
struct inode *ip;
- loff_t off, i_size;
- size_t len, wrote;
- cred_t *cr;
- void *pb;
+ struct page *pl[1];
int error = 0;
ASSERT(PageLocked(pp));
ip = pp->mapping->host;
- off = page_offset(pp);
- i_size = i_size_read(ip);
- ASSERT3S(off, <, i_size);
+ pl[0] = pp;
- cr = (cred_t *)get_current_cred();
- len = MIN(PAGE_CACHE_SIZE, i_size - off);
+ error = -zfs_getpage(ip, pl, 1);
- pb = kmap(pp);
+ if (error) {
+ SetPageError(pp);
+ ClearPageUptodate(pp);
+ } else {
+ ClearPageError(pp);
+ SetPageUptodate(pp);
+ flush_dcache_page(pp);
+ }
- /* O_DIRECT is passed to bypass the page cache and avoid deadlock. */
- wrote = zpl_read_common(ip, pb, len, off, UIO_SYSSPACE, O_DIRECT, cr);
- if (wrote != len)
- error = -EIO;
+ unlock_page(pp);
+ return error;
+}
- if (!error && (len < PAGE_CACHE_SIZE))
- memset(pb + len, 0, PAGE_CACHE_SIZE - len);
+int
+zpl_putpage(struct page *pp, struct writeback_control *wbc, void *data)
+{
+ int error;
- kunmap(pp);
- put_cred(cr);
+ /*
+ * Disable the normal reclaim path for zpl_putpage(). This
+ * ensures that all memory allocations under this call path
+ * will never enter direct reclaim. If this were to happen
+ * the VM might try to write out additional pages by calling
+ * zpl_putpage() again resulting in a deadlock.
+ */
+ current->flags |= PF_MEMALLOC;
+ error = -zfs_putpage(pp, wbc, data);
+ current->flags &= ~PF_MEMALLOC;
if (error) {
SetPageError(pp);
}
unlock_page(pp);
+ return error;
+}
- return (error);
+static int
+zpl_writepages(struct address_space *mapping, struct writeback_control *wbc)
+{
+ return write_cache_pages(mapping, wbc, zpl_putpage, mapping);
}
/*
* support mmap(2). Mapped pages may be dirtied by memory operations
* which never call .write(). These dirty pages are kept in sync with
* the ARC buffers via this hook.
- *
- * Currently this function relies on zpl_write_common() and the O_DIRECT
- * flag to push out the page. This works but the more correct way is
- * to update zfs_putapage() to be Linux friendly and use that interface.
*/
static int
zpl_writepage(struct page *pp, struct writeback_control *wbc)
{
- struct inode *ip;
- loff_t off, i_size;
- size_t len, read;
- cred_t *cr;
- void *pb;
- int error = 0;
-
- ASSERT(PageLocked(pp));
- ip = pp->mapping->host;
- off = page_offset(pp);
- i_size = i_size_read(ip);
-
- cr = (cred_t *)get_current_cred();
- len = MIN(PAGE_CACHE_SIZE, i_size - off);
-
- pb = kmap(pp);
-
- /* O_DIRECT is passed to bypass the page cache and avoid deadlock. */
- read = zpl_write_common(ip, pb, len, off, UIO_SYSSPACE, O_DIRECT, cr);
- if (read != len)
- error = -EIO;
-
- kunmap(pp);
- put_cred(cr);
-
- if (error) {
- SetPageError(pp);
- ClearPageUptodate(pp);
- } else {
- ClearPageError(pp);
- SetPageUptodate(pp);
- }
-
- unlock_page(pp);
-
- return (error);
+ return zpl_putpage(pp, wbc, pp->mapping);
}
const struct address_space_operations zpl_address_space_operations = {
+ .readpages = zpl_readpages,
.readpage = zpl_readpage,
.writepage = zpl_writepage,
+ .writepages = zpl_writepages,
};
const struct file_operations zpl_file_operations = {
- .open = generic_file_open,
+ .open = zpl_open,
+ .release = zpl_release,
.llseek = generic_file_llseek,
.read = zpl_read,
.write = zpl_write,